Power and Performance Validation Engineer
Graphcore is a Bristol-based AI chipmaker (IPU accelerators), a SoftBank subsidiary and a Molten Ventures portfolio company.
Maintainer signals as of 8/20/2026
About Graphcore
Graphcore (graphcore.ai) is a British semiconductor company building Intelligence Processing Units (IPUs) for AI workloads. Founded in Bristol in 2016, it was acquired by SoftBank in 2024. It is a portfolio company of Molten Ventures.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will lead power, thermal and performance validation for AI compute silicon and systems. You will develop validation plans and Python automation, lead post-silicon bring-up, design benchmarks, analyze performance data, debug complex hardware-software interactions, define validation criteria, and guide technical improvements across validation activities.
Requirements
- Strong experience in power and performance validation, silicon characterization or system performance engineering
- Deep understanding of modern SoC architecture
- Strong Linux systems knowledge
- Strong Python programming skills
- Experience with benchmarking and profiling tools such as stress-ng, fio, perf or iperf
- Experience debugging complex hardware and software interactions
- Ability to define validation methodologies, workload models and test strategies
- Strong analytical, communication and cross-functional collaboration skills
- Experience with AI accelerators, GPUs or high-performance computing is desirable
- C/C++ programming experience is desirable
Responsibilities
- Define and lead validation strategies for power, thermal and performance characterization
- Plan and execute validation activities across projects and milestones
- Develop validation plans for functional, stress, workload and corner-case scenarios
- Lead post-silicon bring-up and characterization
- Validate CPU, memory, interconnect and high-speed I/O subsystems
- Develop Python automation frameworks, regression infrastructure and reporting tools
- Design and execute benchmark workloads, parameter sweeps and performance experiments
- Collaborate with architecture, RTL, firmware, software and systems teams to resolve technical issues
- Define validation metrics, pass/fail criteria and reporting methodologies
- Guide development of workload generators and micro-benchmarks
- Analyze power, thermal and performance data
- Communicate findings, project status, risks and recommendations